UltraSight AI Guidance

K223347

UltraSight Inc. · cleared 2023-07-24 · product code QJU · Radiology

Premarket evidence — what FDA accepted

Device typesamd

source quote (p.4)

“UltraSight AI Guidance is a mobile application based on machine learning that uses artificial intelligence (AI) to provide dynamic real-time guidance on the position and orientation of the transducer to help non-expert users acquire diagnostic-quality tomographic views of the heart.”

AlgorithmDeep Learning Based Algorithm

source quote (p.9)

“Deep Learning Based Algorithm: Yes”

Adaptive (vs locked)FDA source did not state this

PCCPFDA source did not state this

Cybersecurity addressedFDA source did not state this

Validation studies (6)

Bench

sample size not stated

standards: IEC 62304, FDA Guidance on General Principles of Software Validation, January 11, 2002

Standalone

n=312 cases

endpoints: classification performance between “diagnosable” and “non diagnosable" clips of each view

Standalone

n=75 patients

endpoints: view detection classification that distinguishes between three system states: "Hold Position", "Navigate", and "No Heart"

Standalone

n=75 patients

endpoints: frame level accuracy of each guidance cue prediction

Prospective clinical

n=61 patients · 1 site(s)

endpoints: qualitative visual assessment of left ventricular (LV) size; LV function; right ventricular (RV) size; presence of nontrivial pericardial effusion; qualitative assessment of RV function; left atrium size; structural assessment of the aortic, mitral, and tricuspid valves; qualitative assessment of IVC size; diagnostic quality score of 3 of higher based on American College of Emergency Physicians (ACEP) scale

standards: American College of Emergency Physicians (ACEP) scale

Prospective clinical

n=240 patients

Reported performance (4 observations)

aurocas written: “auc”0.988CI [0.985, 0.990]

source quote (p.11)

“The mean AUC was 0.988 with 95% CI [0.985, 0.990] showing good classification performance, relative to the success criteria of AUC > 0.8.”

ppvas written: “PPV (Quality Bar)”0.93CI [0.92, 0.94]

source quote (p.11)

“The mean PPV was 0.93 with 95% CI [0.92, 0.94] relative to the success criteria of PPV > 0.75, showing good classification performance.”

aurocas written: “AUC (Quality Bar)”0.86CI [0.85, 0.87]

source quote (p.11)

“The mean AUC was 0.86 with 95% CI [0.85, 0.87] showing good classification performance, relative to the success criteria of AUC > 0.8.”

aurocas written: “AUC (Probe Guidance)”0.821CI [0.813, 0.827]

source quote (p.11)

“The mean AUC was 0.821 with 95% CI [0.813, 0.827] showing good classification performance, relative to the success criteria of AUC > 0.8.”

Each value carries its own analysis unit and task — never compare or pool across devices. Source: 510(k) summary PDF.

Predicate network

Postmarket — what happened after clearance

recalls in product code, 24mo

MAUDE reports in code, 12mo

—

vs code's own 3-yr baseline

drift signals on this device

re_clearance
The FDA AI/ML device list shows a newer 510(k) K251416 (decision 2025-12-17) from Ultrasight , Ltd. for a matching device line ("UltraSight Guidance") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K251416

Recall and MAUDE counts are product-code-level (reports aren't reliably attributable to one device). Signals are descriptive observables with sources — never a judgment that the device is unsafe or drifting. Snapshot 2026-07-08.

Reimbursement — how devices like this got paid

Not yet tracked — no payment pathway indexed for this clearance (the reimbursement corpus is a growing seed set).